Yi base models converted to Llama architecture. • 4 items • Updated • 9
01-ai/Yi-34B with tensors renamed to match standard Llama modelling code. Model can be loaded without trust_remote_code, but the tokenizer can not.
The branch llama-tokenizer uses the Llama tokenizer class as well.
- Downloads last month
- 85
Safetensors
Model size
34B params
Tensor type
BF16
·